Myristoylation is required for mono-ubiquitination of Gpa1. A, anti-Gpa1 immunoblot of strain BY4741 transformed with an overexpression plasmid (pAD4M) containing either wild type GPA1 (WT), gpa1^G2A (G2A), gpa1^S200A (S200A), or the double mutant as indicated. B, DIC and fluorescence microscopy images of BY4741 transformed with an integrating plasmid (pRS406) containing the native GPA1 promoter and either wild type GPA1 or the indicated mutant fused to the yeast-enhanced GFP (left). C, pheromone-induced growth inhibition assay of strain (ste7Δ gpa1Δ) transformed with single copy plasmids pRS315-STE7 and either pRS316-GPA1 or pRS316-GPA1^S200A, with each gene under the control of its native promoter. Transformed cells were plated onto solid medium and exposed to paper discs containing α-factor pheromone (clockwise from left: 1.5, 4.5, 15, and 45 μg). D, cells expressing FUS1-lacZ reporter and overexpressing (pAD4M) Gpa1 or Gpa1^S200A treated with the indicated concentrations of mating pheromone for 90 min. Results are the mean ± S.E. for three individual experiments each performed in triplicate. E and F, a proteasomal protease-defective mutant (cim3-1), a vacuolar protease-deficient mutant (pep4Δ), or the corresponding isogenic wild type strains transformed with plasmid pAD4M-GPA1, pAD4M-gpa1^S200A, or pAD4M-gpa1^G2A and analyzed by immunoblotting with anti-Gpa1 antibodies. Note that all lanes shown within a single panel are from a single gel and are treated identically through all stages of image acquisition. Poly-ubiquitinated Gpa1 (Gpa1-(Ub)_n), mono-ubiquitinated Gpa1 (Gpa1-Ub), myristoylated, and non-myristoylated Gpa1 (Gpa1) and common nonspecific bands (^*) are indicated. Top and bottom panels are identical except for exposure time.

An N-terminal polybasic stretch can substitute for Gpa1 myristoylation. A, table of Gpa1 N-terminal mutants and their corresponding sequences. B, strain YGS5 expressing single copy (pRS316) Gpa1 (WT), Gpa1^G2A (G2A), or Gpa1^G2A/4K (G2A/4K) were grown at 34 °C to saturation, spotted in serial 2-fold dilutions, and incubated either at 34 °C or room temperature (RT). C, DIC and GFP fluorescence of strain BY4741 expressing integrated (pRS406) Gpa1-GFP, Gpa1^G2A-GFP, or Gpa1^G2A/4K-GFP. D, strain (ste7Δ gpa1Δ) transformed with single copy plasmids pRS315-STE7 and either pRS316-GPA1 or pRS316-GPA1^G2A/4K analyzed with the pheromone-induced growth inhibition assay as described above. E, BY4741 cells expressing FUS1-lacZ reporter and overexpressing (pAD4M) Gpa1 or Gpa1^G2A/4K treated with the indicated concentrations of mating pheromone for 90 min. Results are the mean ± S.E. for four individual experiments each performed in triplicate.

Plasma membrane localization is not sufficient for Gpa1 mono-ubiquitination. pAD4M-GPA1 (WT) or the indicated point mutant expressed in pep4Δ or the corresponding isogenic wild type strain and analyzed by immunoblotting with anti-Gpa1 antibodies, as described above. Top and bottom panels are identical except for exposure time. The slower mobility of G2A/4K relative to G2A is presumed to occur because of the effect of substituting four lysine residues on SDS-PAGE mobility of the protein.

Plasma membrane localization is sufficient for Gpa1 poly-ubiquitination. pAD4M-GPA1 (WT) or the indicated point mutant expressed in cim3-1 or the corresponding isogenic wild type strain and analyzed by immunoblotting with anti-Gpa1 antibodies as described above. Quantitation of the poly-ubiquitinated form of Gpa1 (laddering) relative to native Gpa1 is the average ± S.D. for three independent experiments. Top and bottom panels are identical except for exposure time.

Gpa1 poly-ubiquitination, but not mono-ubiquitination, is elevated for protein folding mutants. A, immunoblot of mutant cim3-1 and the isogenic wild type strain transformed with plasmid pAD4M-GPA1 (WT) or the indicated point mutant grown at the restrictive temperature of 37 °C for 4 h. Whole cell extracts were resolved by 10% SDS-PAGE and immunoblotting with anti-Gpa1 antibodies. B, mutant pep4Δ and the isogenic wild type strain transformed with plasmid pAD4M-GPA1 or the indicated point mutant and grown to early-log phase at 30 °C. The difference in exposures between cim3-1 and pep4Δ experiments is required to ensure visualization of either the mono-ubiquitin band (which is in close proximity to the Gpa1 bands) or poly-ubiquitin laddering. Quantitation of the poly-ubiquitinated (cim3-1 strain) or mono-ubiquitinated (pep4Δ strain) Gpa1 relative to native Gpa1 is the average ± S.D. for at least three independent experiments. C, ste7Δ gpa1Δ cells expressing pRS315-STE7 and pRS316-GPA1 or the indicated mutant were grown to early log phase and analyzed by immunoblotting with anti-Gpa1 antibodies.

The ubiquitin ligase Rsp5 is required for Gpa1 mono-ubiquitination in vivo. A, cells harboring a doxycycline-repressible promoter attached to the RSP5 open reading frame (TetO₇-RSP5) or to a non-expressible genetic element (TetO₇-WT) were transformed with the overexpression plasmid pAD4M-GPA1 or empty vector. Transformants were grown to early log phase and then split, and half were treated with doxycycline (Dox) at 10 μg/ml for 16 h, diluted in doxycycline-containing or free medium, and grown to early log phase. Cells were harvested, lysed, and analyzed by immunoblotting with anti-Gpa1 antibodies. B, DIC and GFP fluorescence of strain TetO₇-WT or TetO₇-RSP5 harboring pRS405-GPA1-GFP-pep4 integrated into the PEP4 locus, thereby disrupting expression of the vacuolar protease Pep4. C, mono-ubiquitination of overexpressed wild type (WT) and non-myristoylated (G2A) Gpa1 was compared by Gpa1 immunoblotting with that of a Gpa1 double mutant P407A/Y409A (ΔPY) lacking the essential residues of a putative Rsp5 binding motif (PY motif) (left). Quantitation of the resulting immunoblots was performed in triplicate using cells from three different transformants.

Rsp5 ubiquitinates Gpa1 in vitro. Purified recombinant wild type or catalytically-inactive Rsp5 (Rsp5-(C/A)), Gpa1, and Uba1 (E1) were incubated as indicated in the presence of Ubc5, ATP, and ubiquitin for 30 min at room temperature followed by quenching with SDS-PAGE loading buffer containing DTT. Samples were resolved by 10% SDS-PAGE and immunoblotting with anti-Gpa1 antibody (top panels). The nitrocellulose blot was stained with Coomassie Blue to detect Rsp5 and Uba1.

Determinants of Gpa1 mono- and poly-ubiquitination. The fully myristoylated and fully mature form of Gpa1 (e.g. Gpa1^S200A) is mono-ubiquitinated and translocated to the vacuole for degradation (enriched in pep4Δ cells). Non-myristoylated (e.g. Gpa1^G2A or Gpa1^G2A/4K) or misfolded Gpa1 is poly-ubiquitinated and degraded by the proteasome. Rsp5 is necessary and sufficient for Gpa1 mono-ubiquitination. The enzyme responsible for Gpa1 poly-ubiquitination is unknown. PM, plasma membrane.